Advances and Applications in Statistics
Volume 3, Issue 1, Pages 15 - 32
(April 2003)
|
|
AUTOMATIC CONTINUOUS SMOOTHING
Erhard Reschenhofer (Austria) and Lakdere Benkherouf (Oman)
|
Abstract: This paper proposes an
efficient method for automatic continuous
smoothing, which is based on the determination
of the trade-off between the goodness of fit and
the danger of over-fitting. The usefulness of
this method is demonstrated by applying it to
economic time series. It is also shown how this
method can be used for spectral density
estimation. Algorithms are given for the
automatic selection of the degree of smoothing
as well as for the carrying out of the
smoothing. |
Keywords and phrases: penalized least squares estimators, Hodrick-Prescott filter, model selection criteria, nonparametric spectral density estimation. |
|
Number of Downloads: 540 | Number of Views: 1521 |
|